FIELD OF THE INVENTION
The present invention relates to a semiconductor device that includes a CMOS transistor having a salicide structure, and more particularly it relates to a semiconductor device having a means for detecting a missing part in a high-melting-point metal silicide film in the connection part between an n-channel MOS transistor gate electrode and a p-channel MOS transistor gate electrode.
RELATED ART
In recent years, with shrinking of the design rule for MOS transistors, because it has become essential to reduce the parasitic resistance associated with the gate electrode, MOS transistors having a salicide structure have come into focus. In these MOS transistors having a salicide structure, with advancements in the shrinking of the design rule, it has become difficult to form a high-melting-point silicide layer in a self-aligning manner on a polysilicon film pattern which serves as the gate electrode, and on a diffusion layer which serves as a source/drain region, and particularly, missing parts occur in the high-melting-point silicide film pattern and high-melting-point metal silicide layer.
A detection means for detecting such missing parts is one in which a detection element formed by a n+-type polysilicon film pattern of sufficient length or p+-type polysilicon film pattern and a high-melting-point metal silicide film pattern formed in self-aligning manner thereover so as to form a polycide interconnect pattern, or a detection element formed by a n+-type diffusion layer or p+-type diffusion layer of sufficient length over which a high-melting-point silicide layer is formed in self-aligning manner is used, the resistance of either of these detection elements being measured to perform detection of missing parts. If there is no missing part in the high-melting-point metal silicide film pattern or the high-melting-point metal silicide layer, the above-noted resistance value is a low value. For example, in the case in which there is a macro-level missing part having a length on the order of micron in the high-melting-point metal silicide film pattern or high-melting-point metal silicide layer, the above-noted resistance value is high.
For example, in the case of a CMOS transistor having a salicide structure, the inverter gate electrode is as follows. In the element-separation regions of both transistors, an n+-type polysilicon film pattern (forming an n-channel MOS transistor gate electrode) and a p+-type polysilicon film pattern (forming the p-channel MOS transistor gate electrode) are directly connected, and a high-melting-point metal silicide film pattern is formed in self-aligning manner over these polysilicon film patterns, so as to constitute a gate electrode of the inverter. A missing part in the high-melting-point metal silicide film pattern on the inverter gate electrode tends to occur at the joining point between the n+-type polysilicon film pattern and the p+-type polysilicon film pattern, and the length of this missing part is in the order of no more than sub-micron size. When this type of missing part occurs on this p+-n+ junction, because the existence of this p+-n+ junction causes an increase in resistance, thereby causes an increase in the inverter gate delay time.
However, in detecting a missing part in a high melting point metal silicide film using a detection element such as described above, because there is only a small increase in resistance value caused by the existence of a missing part having a length in the sub-micron order, this approach to detecting a missing part at the junction of a n+-type polysilicon film pattern and a p+-type polysilicon film pattern formed on the gate electrode of the inverter is not suitable.
Accordingly, it is an object of the present invention to provide a semiconductor device, which includes a high-melting-point metal silicide film missing part detection element capable of detecting a minute missing part in a high-melting-point metal silicide film of a polycide interconnect in a semiconductor device including a CMOS transistor of salicide structure.
SUMMARY OF THE INVENTION
In order to achieve the above-noted object, the present invention adopts the following basic technical constitution.
The first aspect of the present invention is a semiconductor device comprising an n-channel MOS transistor 1 and a p-channel MOS transistor 2 each having a salicide structure, and a high-melting point metal silicide film missing part detection element 3 detecting a missing part 128 formed in a high-melting point metal silicide film 126A provided on a gate electrode 106 a, 106 b of the MOS transistors 1,2, wherein the high-melting point metal silicide film missing part detection element 3 comprising a polysilicon film pattern 110 with alternate connection between an n-type region 116 ab having a prescribed impurity concentration and a p-type region 116 ba having a prescribed impurity concentration, and a high-melting point metal silicide film 126A formed on the polysilicon film pattern 110.
In the second aspect of the present invention, the impurity concentration of the p-type region is higher than that of the n-type region.
In the third aspect of the present invention, both ends of the polysilicon film pattern are made of the p-type region.
In the fourth aspect of the present invention, the impurity concentration of the n-type region is higher than that of the p-type region.
In the fifth aspect of the present invention, both ends of the polysilicon film pattern are made of the n-type region.
In the sixth aspect of the present invention, a source/drain region of the MOS transistor is formed by a LDD region and a source/drain diffusion layer having an impurity concentration higher than that of the LDD region, an impurity concentration of n-type region of the high-melting point metal silicide film missing part detection element is equal to that of the source/drain diffusion layer of the n-channel MOS transistor, and an impurity concentration of p-type region of the high-melting point metal silicide film missing part detection element is equal to that of the source/drain diffusion layer of the p-channel MOS transistor.
In the seventh aspect of the present invention, a source/drain region of the MOS transistor is formed by a LDD region and a source/drain diffusion layer having an impurity concentration higher than that of the LDD region, an impurity concentration of n-type region of the high-melting point metal silicide film missing part detection element is equal to that of the LDD region of the n-channel MOS transistor.
In the eighth aspect of the present invention, a source/drain region of the MOS transistor is formed by a LDD region and a source/drain diffusion layer having an impurity concentration higher than that of the LDD region, an impurity concentration of p-type region of the high-melting point metal silicide film missing part detection element is equal to that of the LDD region of the p-channel MOS transistor.
In the ninth aspect of the present invention, minimum line width of the polysilicon film pattern of the high-melting point metal silicide film missing part detection element is equal to a gate length of a gate electrode of the MOS transistor.
In the tenth aspect of the present invention, the high-melting point metal is either one of titanium or cobalt.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1(a) is a schematic cross-sectional view showing a high-melting point metal silicide film missing part detection element according to the first embodiment of the present invention.
FIG. 1(b) is a schematic cross-sectional view showing a CMOS transistor having a salicide structure according to the first embodiment of the present invention.
FIG. 2(a) to FIG. 2(d) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the first embodiment of the present invention.
FIG. 3(a) to FIG. 3(c) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the first embodiment of the present invention.
FIG. 4(a) to FIG. 4(d) are is schematic cross-sectional views showing the manufacturing processes for a CMOS transistor according to the first embodiment of the present invention.
FIG. 5(a) to FIG. 5(c) are is schematic cross-sectional views showing the manufacturing processes for a CMOS transistor according to the first embodiment of the present invention.
FIG. 6(a) is a schematic cross-sectional view showing a high-melting point metal silicide film missing part detection element according to the second embodiment of the present invention.
FIG. 6(b) is a schematic cross-sectional view showing a CMOS transistor having a salicide structure according to the second embodiment of the present invention.
FIG. 7(a) to FIG. 7(c) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the second embodiment of the present invention.
FIG. 8(a) is a schematic cross-sectional view showing a high-melting point metal silicide film missing part detection element according to the third embodiment of the present invention.
FIG. 8(b) is a schematic cross-sectional view showing a CMOS transistor having a salicide structure according to the third embodiment of the present invention.
FIG. 9(a) to FIG. 9(d) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the third embodiment of the present invention.
FIG. 10 (a) to FIG. 10(c) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the third embodiment of the present invention.
FIG. 11(a) to FIG. 11(d) are is schematic cross-sectional views showing the manufacturing processes for a CMOS transistor according to the third embodiment of the present invention.
FIG. 12(a) to FIG. 12(c) are is schematic cross-sectional views showing the manufacturing processes for a CMOS transistor according to the third embodiment of the present invention.
FIG. 13(a) is a schematic cross-sectional view showing a high-melting point metal silicide film missing part detection element according to the fourth embodiment of the present invention.
FIG. 13(b) is a schematic cross-sectional view showing a CMOS transistor having a salicide structure according to the fourth embodiment of the present invention.
FIG. 14(a) to FIG. 14(d) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the fourth embodiment of the present invention.
FIG. 15(a) to FIG. 15(c) are schematic cross-sectional views showing the manufacturing processes for a high-melting point metal silicide film missing part detection element according to the fourth embodiment of the present invention.
FIG. 16(a) to FIG. 16(d) are is schematic cross-sectional views showing the manufacturing processes for a CMOS transistor according to the fourth embodiment of the present invention.
FIG. 17(a) to FIG. 17(c) are is schematic cross-sectional views showing the manufacturing processes for a CMOS transistor according to the fourth embodiment of the present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Preferred embodiments of the present invention are described in detailed below, with references made to relevant accompanying drawings.
First Embodiment
Specifically, the configuration of a semiconductor device according to a first embodiment of the present invention will be described with reference to FIG. 1(b), which is a schematic cross-sectional view of a CMOS transistor having a salicide structure, and FIG. 1(a), which is a schematic cross-sectional view of a high-melting-point metal silicide film missing part detection element.
Onto the surface of a silicon substrate 101, a p-type well 102 and n-type well 103, and a field insulation film 104 are provided. An element-separation region provided on the surface of the silicon substrate 101 is, for example, formed with an STI structure, in which a shallow groove is filled with the field insulation film 104. The p-type well 102 and n-type well 103 are separated by this field insulation film 104. An n-channel MOS transistor is provided on the surface of the p-type well 102, and a p-channel MOS transistor is provided on the surface of the n-type well 103, the above-noted high-melting-point metal silicide film missing part detection element being provided on the surface of the field insulation film 104, for example. The n-channel MOS transistor and p-channel MOS transistor, and the high-melting-point metal silicide film missing part detection element are covered by an interlayer insulation film 135.
The n-channel MOS transistor 1 is formed by a gate oxide film 105, a gate electrode 106 a provided on the surface of the p-type well 102 via the gate oxide film 105, and a source/drain region 108 a provided on the surface of the p-type well 102. The side surface of the gate electrode 106 a is directly covered by an insulating film spacer 122. The source/drain region 108 a is formed by an n-type source/drain extension diffusion layer (Lightly Doped Drain, LDD) 120 provided on the surface of the p-type well 102, self-aligned to the gate electrode 106 a, an n+-type source/drain diffusion layer 123 aa provided on the surface of the p-type well 102, self-aligned to the gate electrode 106 a and the insulation film spacer 122, and a titanium silicide (TiSi2) layer 127 aa provided on the surface of the n+-type source/drain diffusion layer 123 aa, self-aligned to the gate electrode 106 a and insulation film spacer 122. The gate electrode 106 a directly covers the surface of the gate oxide film 105, and is formed by a laminate film comprising an n+-type polysilicon film pattern 113 ab extending over the surface of the field insulation film 104 and a titanium silicide film pattern 127 a formed in self-aligning manner over the surface of the n+-type polysilicon film pattern 113 ab.
The minimum line width of the gate electrode 106 a is, for example, 0.2 μm to 0.35 μm, and the film thickness of the titanium silicide film 127 a and titanium silicide film 127 aa is approximately 270 nm to 450 nm, the depth of the junction of the n+-type source/drain diffusion layer 123 aa being, for example, approximately 0.15 μm. The n-type source/drain extension diffusion layer 120 a is formed by ion implantation of arsenic (As) on the order of 1014 cm−2, the n+-type source/drain diffusion layer 123 aa is formed by ion implantation of arsenic at approximately 2×1015 cm−2. The insulation film spacer 122 is, for example, a silicon oxide film having a film thickness of approximately 50 nm or a silicon nitrate film of this thickness. The film thickness of the gate oxide film 105 is, for example, approximately 5 nm to 6 nm.
The p-channel MOS transistor 2 is formed by a gate oxide film 105, a gate electrode 106 b provided on the surface of the n-type well 103, via the gate oxide film 105, and a source/drain region 108 b provided on the surface of the n-type well 103. The side surface of the gate electrode 106 b as well is directly covered by the insulation film spacer 122. The source/drain region 108 b is formed by a p-type source/drain extension diffusion layer 120 b provided on the surface of the n-type well 103, self-aligned to the p+-type polysilicon film pattern 113 bb, a p+-type source/drain diffusion region 123 ba provided on the surface of the n-type well 103, self-aligned to the gate electrode 106 b and the insulation film spacer 122, and a titanium silicide layer 127 ba formed on the surface of a p+-type source/drain diffusion layer, self-aligned to the gate electrode 106 b and insulation film spacer 122. The gate electrode 106 b directly covers the surface of the gate oxide film 105, and is formed by a p+-type polysilicon film pattern 113 bb extending over the surface of the field insulation film 104 and a titanium silicide film pattern 127 b provided self-aligned on the surface of the p+-type polysilicon film pattern 113 bb.
The minimum line width of the gate electrode 106 b is the same as the minimum line width of the gate electrode 106 a. The film thicknesses of the titanium silicide film 127 b and titanium silicide layer 127 ba are approximately 10% thicker than that of the titanium silicide film 127 a and titanium silicide layer 127 aa, this being for example 300 nm to 500 nm. The depth of the junction of the p+-type source/drain diffusion layer 127 ba is, for example, approximately 0.2 μm. The p-type source/drain extension diffusion layer 120 b is formed by ion implantation of boron (B) or boron fluoride (BF2) to a concentration on the order of 1014 cm−2. The p+-type source/drain diffusion layer 123 ba is formed by ion implantation of boron at approximately 5×1015 cm−2, at an energy level of, for example, 2 keV.
Contact holes that reaches the source/ drain regions 108 a and 108 b is provided in the interlayer insulation film 135. A metal interconnect connected to one side of the source/drain region 108 a, a metal inter connect 137 b connected to one side of the source/drain region 108 b and a metal interconnect 137 c that is connected to the other the source/drain region 108 a and the other source/drain 108 b are provided on the surface of the interlayer insulation film 135, via the contact holes. For example, the metal interconnect 137 a is connected to the power supply voltage (VDD), and the metal interconnect 137 b is connected to the ground voltage (GND).
As shown in FIG. 1(a), the high-melting-point metal silicide film missing part detection element 3 is formed, for example, by a polysilicon film pattern 110 that directly covers the surface of the field insulation film 104, and a titanium silicide film pattern 126A provided on the surface of the polysilicon film pattern 110 in a self-aligning manner. The minimum line width of the high-melting-point metal silicide film missing part detection element is the same as the minimum line width the gate electrode 106 a. The side surface of the high-melting-point metal silicide film missing part detection element is directly covered by the insulation film spacer 122. In this first embodiment of the present invention, the high-melting-point metal silicide film missing part detection element is not restricted to being provided on the surface of the field insulation film 104, and can alternately be provided, for example, so as to directly cover the surface of the gate oxide film 105.
The polysilicon film pattern 110 is formed by the mutual connection between an n+-type region 116 ab and the p+-type type region 116 ba. The lengths of the n+-type type region 116 ab and p+-type region 116 ba are each approximately 0.5 μm. The film thickness of the titanium silicide film pattern 126A is not uniform, and, in the part at the upper surface of the n+-type region 116 ab, and p+-type region 116 ba, it is the same as the film thicknesses of the titanium silicide film patterns 127 a and 127 b, respectively. Directly above the junction part between the n+-type region 116 ab and the p+-type region 116 ba, it is possible that a missing part 128 occurs in the titanium silicide film pattern 126A. The mechanism for the occurrence of the missing part 128 is described below.
A contact hole that reaches the two end parts of the high-melting-point metal silicide film missing part detection element is provided in the interlayer insulation film 135. A pair of metal interconnects 136 connected to each of the ends of the high-melting-point metal silicide film missing part detection element is provided on the surface of the interlayer insulation film 135, via these contact holes. For this reason, the line width at the two ends of the high-melting-point metal silicide film missing part detection element is wider than the minimum line width thereof (for example, approximately 0.5 μm. In the case in which at least one missing part 128 occurs in the titanium silicide film pattern 126A, if a bias is applied between the pairs of metal interconnects 136, and the resistance is measured, the resistance value measured will be a value that is sufficiently higher than the simple sheet resistance value. In the case in which it is necessary to detect the above-noted missing part 128 immediately before the formation of the interlayer insulation form 135, because it is necessary to make connection by a needle probe to these parts, the line width of the two ends of the high-melting-point metal silicide film missing part detection element is, for example, 20 μm or larger.
Specifically, in the first embodiment of the present invention, in a semiconductor device that includes a CMOS transistor having a salicide structure, a high-melting-point metal silicide film missing part detection element formed by the polysilicon film pattern 110 and the titanium silicide film pattern 126A is provided, so that it is easy to detect a minute missing part in the high-melting-point metal silicide film in the gate electrode made of polycide interconnect.
Referring to FIG. 2 and FIG. 3 which are schematic cross-sectional views of the process for manufacturing a high-melting-point metal silicide film missing part detection element, FIG. 4 and FIG. 5, which are schematic cross-sectional views showing the process for manufacturing a titanium salicide structured CMOS transistor, and the above-noted FIG. 1 are used below to further describe the first embodiment of a semiconductor device according to the present invention.
First, a p-type well 102, an n-type well 103, and a field insulation film 104 formed by a silicon oxide insulation film are formed on the surface of a silicon substrate 101. This field insulation film 104 provides element separation between the p-type well 102 and the n-type well 103. This field insulation film 104 is formed, for example, by filling a shallow groove, so as to form an element separation region having an STI structure on the surface on the silicon substrate 101. A gate oxide film 105 having a film thickness of, for example, 5 nm to 6 nm is formed by thermal oxidation on the surface of the p-type well 102 and n-type well 103.
Next, using, for example, an LPCVD process, an undoped polysilicon film (not shown in the drawing) having a film thickness of 150 nm to 200 nm is formed over the entire surface of the substrate. This undoped polysilicon film is patterned, and the undoped polysilicon film pattern 112 of the high-melting-point metal silicide film missing part detection element is formed on the surface of the field insulation film 104. Simultaneously, the undoped polysilicon film pattern 113 for the gate electrode on the gate electrode oxide 105 is formed, for example, extending from the surface of the p-type well 102, via the surface of the field insulation film 104, to the surface of the n-type well 103. The minimum line width of the undoped polysilicon film pattern 112 is the same as the minimum line width (gate length) of the undoped polysilicon film pattern 113, this being, for example, 0.2 μm to 0.35 μm. The line width of both ends of the undoped polysilicon film pattern 112 is made wider than this minimum line width, and is approximately, for example, 0.5 μm, and can be made 20 μm or greater, if necessary.
Next, a first photoresist film pattern 141 having an aperture at the part corresponding to the p-type well 102 and covering the undoped polysilicon pattern 112 the surface of the n-type well 103 is formed. This photoresist film pattern 141 is used as a mask to perform ion implantation of arsenic to the order of 1014 cm−2, thereby part of the undoped polysilicon film pattern 113 becomes an n-type polysilicon film pattern 113 a and simultaneously an n-type source/drain extension diffusion layer 120 a self-aligned to the n-type polysilicon film pattern 113 a, is formed on the surface of the p-type well 102. (refer to FIG. 2(a) and FIG. 4(a)).
After removing the above-noted photoresist film pattern 141, a second photoresist film pattern 142 is formed. This photoresist film pattern 142 is formed so as to have an aperture at a part corresponding to the n-type well 103 and so as to cover the undoped polysilicon film pattern 112 and the surface of the p-type well 102. This photoresist film pattern 142 is used as a mask to perform ion implantation of boron (or boron fluoride) to the order of 1014 cm−2, so that the remaining part of the undoped polysilicon film pattern 113 becomes as a p-type polysilicon film pattern 113 b, simultaneously with which a p-type source/drain extension diffusion layer 120 b is formed on the surface of the n-type well 103, self-aligned to the p-type polysilicon film pattern 113 b (refer to FIG. 2(b) and FIG. 4(b)).
It will be understood that in the first embodiment of the present invention there is no restriction with regard to the sequence of steps related to the photoresist film patterns 141 and 142, and that they can alternately be reversed from the sequence presented above. Additionally, if necessary, it is possible to perform ion implantation of a p-type impurity and n-type impurity, using the photoresist film pattern 141 and 142 so as to form, respectively, a p-type pocket diffusion layer and an n-type pocket diffusion layer on the surfaces of the p-type well 102 and the n-type well 103.
Next, the photoresist film pattern 142 is removed, and an LPCVD process is used to deposit an insulation film (silicon oxide or silicon nitride) having a film thickness of approximately 50 nm, for example, over the entire surface. This insulation film is etched back using anisotropic etching, so as to form an insulation spacer 122 on the side of the undoped polysilicon film pattern 112, the n-type polysilicon film pattern 113 a, and the p-type polysilicon film pattern 113 b, respectively.
Next, a third photoresist film pattern 143 is formed. The photoresist film pattern 143 has an aperture at a part corresponding to the p-type well 102 and minimally covers the surface of the n-type well 103. Additionally this photoresist film pattern 143 has a plurality of apertures at a part that covers the undoped polysilicon film pattern 112. The aperture width and spacing of the apertures in the part that covers the undoped polysilicon film pattern 112 (the part other than the end part of the undoped polysilicon film pattern 112) are each, for example, approximately 0.5 μm. This photoresist film pattern 143 is used as a mask, for example, to perform ion implantation of arsenic to approximately 2×1015 cm−2, 11 so that the n-type polysilicon film pattern 113 a becomes an n+-type polysilicon film pattern 113 aa and so that an n+-type source/drain diffusion region diffusion layer 123 a is formed so as to be self-aligned with then+-type polysilicon film pattern 113 aa and insulation film spacer 122 on the surface on the p-type well 102. Simultaneously with this, an n+-type region 116 a is formed on the undoped polysilicon film pattern 112. The depth of the junction of the n+-type source/drain diffusion layer 123 a is, for example, approximately 0.15 μm and the length and spacing of the n+-type region 116 a, are under 0.6 μm and greater than 0.4 μm, respectively (refer to FIG. 2(c) and FIG. 4(c)).
Next, after removing the above-noted photoresist film pattern 143, a fourth photoresist film pattern 144 is formed. The photoresist film pattern 144 has an aperture at a part corresponding to the n-type well 103 and minimally covers the p-type well 102 surface. Additionally, this photoresist film pattern 144 is provided with a plurality of apertures in a part in which the undoped polysilicon film pattern 112 has not been doped to become an n+-type region 116 a. The aperture width and spacing in this part (the part other than the end part of the undoped polysilicon film pattern 112) are, for example, 0.5 μm. This photoresist film pattern 144 is used as a mask to perform ion implantation of boron to approximately 5×1015 cm−2 using an energy of, for example, 2 keV, so that the p-type polysilicon film pattern 113 b becomes a p+-type polysilicon film pattern 113 ba, and a p+-type source/drain diffusion layer 123 b is formed on the surface of the n-type well 103, self-aligned to the p+-type polysilicon film pattern 113 ba and the insulation films spacer 122. Simultaneously with this, in the part of the undoped polysilicon film pattern 112 in which the n+-type region 116 a is not formed, a p+-type region 116 b is formed and the n+-type region 116 a becomes the n+-type region 116 aa. The depth of the junction the p+-type source/drain diffusion layer 123 b is, for example, approximately 0.2 μm, and the lengths of the n+-type region 116 aa and p+-type region 116 b, are each approximately, for example, 0.5 μm (refer to FIG. 2(d) and FIG. 4(d)). In the first embodiment, it will be understood that there is no restriction to the sequence of the processed steps related to the photoresist film patterns 143 and 144, and that they can alternately be reversed from the sequence presented above.
Next, after removing the photoresist film pattern 144, sputtering or the like is used to form a titanium film 125 over the entire surface area (refer to FIG. 3(a) and FIG. 5 (a)).
Next, a siliciding reaction is done at a prescribed reaction temperature. By doing this, in a part that forms a high-melting-point silicide film missing part detection element, the n+-type region 116 aa and p+-type region 116 b become the n+-type region 116 ab and p+-type region 116 ba, respectively, and the original undoped polysilicon film pattern 112 becomes a polysilicon film pattern 110 in which there is alternating connection between the n+-type region 116 ab and p+-type 116 ba, a titanium silicide film pattern 126 being formed in self-aligning manner on the surface of the polysilicon film pattern 110.
In the part that becomes the n-channel MOS transistor, the n+-type polysilicon pattern 113 aa becomes the n+-type polysilicon film pattern 113 ab, and a titanium silicide film pattern 127 a is formed in self-aligning manner on the surface of the n+-type polysilicon film pattern 113 ab, a gate electrode 106 a being formed from these laminated semiconductor films. Simultaneously, the n+-type source/drain diffusion layer 123 a becomes the n+-type source/drain diffusion layer 123 aa, and a titanium silicide layer 127 aa is formed in self-aligning manner on the surface of the n+-type source/drain diffusion layer 123 aa, so that a source/drain region 108 a being formed by the n+-type source/drain extension diffusion region 120 a, the n+-type source/drain diffusion region 123 aa, and the titanium silicide layer 127 aa.
In a part that becomes the p-channel MOS transistor, the p+-type polysilicon film pattern 113 ba becomes the p+-type polysilicon film pattern 113 bb and a titanium silicide film pattern 127 b is formed in self-aligning manner on the surface of the p+-type polysilicon film pattern 113 bb, a gate electrode 106 b being formed from this laminated semiconductor film. Simultaneously, the p+-type source/drain diffusion layer 123 b becomes the p+-type source/drain diffusion layer 123 ba, and a titanium silicide layer 127 ba is formed in self-aligning manner on the surface of the p+-type source/drain diffusion layer 123 ba, so that a source/drain diffusion region 108 b being formed by the p-type source/drain extension diffusion layer 120 b, the p+-type source/drain diffusion layer 123 ba, and the titanium silicide layer 127 ba (refer to FIG. 3(b) and FIG. 5(b)).
In the titanium siliciding reaction, it is known that silicon diffuses into the titanium. When this occurs, if arsenic exists in a high concentration within the silicon, the speed of the siliciding reaction is reduced. As a result, the film thickness of the titanium silicide film pattern 126 in the part directly above the n+-type region 116 ab (and also the titanium silicide film 127 a and titanium silicide layer 127 aa) becomes smaller than (approximately 10%) the film thickness of the titanium silicide film pattern 126 directly above the p+-type region 116 ba (and the titanium silicide film 127 b and the titanium silicide layer 127 ba).
Next, the unreacted titanium layer 125 that directly covers the surface of the field insulation film 104 and insulation film spacer 122 is selectively removed (refer to FIG. 3(c) and FIG. 5(c)).
After the above is done, an interlayer insulation film 135 made of a silicon oxide insulation film is formed over the entire surface and a contact hole is formed in this interlayer insulation film 135. Additionally the metal interconnects 136, 137 a, 137 b, and 137 c are formed on the surface of the interlayer insulation film 135, thereby completing the semiconductor device according to the first embodiment of the present invention.
During the manufacturing process including the above-noted siliciding reaction, and after completion of the semiconductor device, a missing part 128 in the titanium silicide film pattern 126 occurs at a position corresponding to a position directly above the joining part between the n+-type region 116 ab and p+-type region 116 ba in the polysilicon film pattern 110, so that the titanium silicide film pattern 126 becomes the titanium silicide film pattern 126A. The occurrence of the missing part 128 is due to the difference between the film thickness of the titanium silicide film pattern above the n+-type region 116 ab and the film-thickness of the titanium silicide film pattern above the p+-type region 116 ba. Furthermore, in the case of titanium silicide, in the low-resistance phase C54 is used in the semiconductor element. For this reason, it is necessary to have a phase transition in the crystal structure from the high-resistance phase C49 to the low-resistance phase C54, and coagulation of the titanium silicide film occurred in the phase transition is also a cause of the tendency for the above-noted missing part 128 to occur.
Second Embodiment
The second embodiment of the present invention will be described with reference to FIG. 6(b) which is a schematic cross-sectional view of a CMOS transistor having a salicide structure, and FIG. 6(a) which is a schematic cross-sectional view of a high-melting-point metal silicide film missing part detection element. The second embodiment differs from the first embodiment in that the semiconductor device according to the second embodiment includes a CMOS transistor having a cobalt salicide structure.
In the above-noted semiconductor device, a p-type well 102, and n-type well 103 and a field insulation film 104 are provided on the surface of a silicon substrate 101. An n-channel MOS transistor is provided on the surface of the p-type well 102 and a p-channel MOS transistor is provided on the surface on the n-channel well 103, the above-noted high-melting-point metal silicide film missing part detection element being provided on the surface, for example, of the field insulation film 104. The n-channel MOS transistor, p-channel MOS transistor and high-melting-point metal silicide film missing part detection element are covered by an interlayer insulation film 135.
The n-channel MOS transistor is formed by a gate oxide film 105, a gate electrode 107 a provided on the surface of the p-type well 102 via the gate oxide film 105, and a source/drain region 109 a provided on the surface of the p-type well 102. The side surface of the gate electrode 107 a is directly covered by an insulation film spacer 122. The source/drain region 109 a is formed by an n-type source/drain extension diffusion layer 121 a provided on the surface of the p-type well 102, self-aligned to the gate electrode 107 a, an n+-type source/drain diffusion region 124 aa provided on the surface of the p-type well 102, self-aligned to the gate electrode 107 a and the insulation film spacer 122, and a cobalt silicide (CoSi2) layer 132 aa formed on a surface of the n+-type source/drain diffusion layer 124 aa, self-aligned to the gate electrode 107 a and insulation film spacer 122. The gate electrode 107 a directly covers the surface of the gate oxide film 105, and is formed by a laminate film made of an n+-type polysilicon film pattern 115 ab extending over the surface of the field insulation film 104 and a cobalt silicide film pattern 132 a provided in self-aligned manner on the surface of the n+-type polysilicon film pattern 115 ab.
The minimum line width of the gate electrode 107 a is, for example, 0.18 μm, and the film thickness of the cobalt silicide film 132 a and cobalt silicide film 132 aa is approximately 250 nm to 350 nm, the depth of the junction of the n+-type source/drain diffusion region 124 aa being, for example, 0.10 μm smaller. The n-type source/drain extension diffusion layer 121 a is formed by ion implantation of arsenic in the order of 1014 cm−2, and n+-type source/drain diffusion layer 124 aa is formed, for example, by ion implantation of arsenic in the order of 5×1015 cm−2. The insulation film spacer 122 is formed by a silicon oxide film or a silicon nitride film having a film thickness of, for example, approximately 50 nm. The gate oxide film 105 thickness is, for example, approximately 2.5 nm to 3.5 nm.
The p-channel MOS transistor is formed by the gate oxide film 105, the gate electrode 107 b formed on the surface of the n-type well 103 via the gate oxide film 105, and a drain/source region 109 b provided on the surface of the n-type well 103. The side surface of the gate electrode 107 b as well is directly covered by the insulation film spacer 122. The source/drain region 109 b is formed by a p-type source/drain extension diffusion layer 121 b provided on the surface of the n-type well 103, self-aligned to the p+-type polysilicon film pattern 115 bb, a p+-type source/drain diffusion layer 124 ba provided on the surface of the n-type well 103, self-aligned to the gate electrode 107 b and the insulation film spacer 122, and a cobalt silicide film 132 ba formed on the surface of the p+-type source/drain diffusion layer 124 ba, self-aligned to the gate electrode 107 b and the insulation film spacer 122. The gate electrode 107 b directly covers the gate oxide film 105, and is formed by the p+-type polysilicon film pattern 115 bb extending over the surface of the field insulation film 104, and a cobalt silicide film pattern 132 b provided in self-aligning manner on the surface of the p+-type polysilicon pattern 115 bb.
The minimum line width of the gate electrode 107 b is the same as the minimum line width of the gate electrode 107 a. The film thicknesses of the cobalt silicide film 132 b and the cobalt silicide layer 132 ba, different than the relationship between the titanium silicide thicknesses over the n+ layer and p+ layer in the first embodiment, are 5% thinner than the film thicknesses of the cobalt silicide film 132 a and the cobalt silicide layer 132 aa, this being for example approximately 240 nm to 330 nm. The thickness of the junction of the p+-type source/drain diffusion layer 124 ba is, for example, approximately 0.15 μm. The p-type source/drain extension diffusion layer 121 b is formed by ion implantation of boron (B) or boron fluoride (BF2) to the order of 1014 cm−2. The p+-type source/drain diffusion layer 124 ba is formed by ion implantation of boron to approximately 2×1015 cm−2 to 3×15 cm−2, using an energy of 2 keV, for example.
A contact hole that reaches the source/ drain regions 109 a and 109 b is provided in the interlayer insulation film 135. Withun these contact holes, a metal interconnect 139 a connected to one side of the source/drain region 139 a, a metal interconnect 139 b connected to one side of the source/drain region 109 b, and a metal interconnect 139 c which is connected to the other source/drain region 109 a and the other source/drain region 109 b are provided. For example, the metal interconnect 139 a is connected to the power supply voltage, and the metal interconnect 139 b is connected to the ground voltage.
The high-melting-point metal silicide film missing part detection element is formed, for example, by a polysilicon film pattern 111 that directly covers the surface of the filled insulation film 104, and a cobalt film silicide pattern 131A provided in the self-aligned manner on the surface of the polysilicon pattern 111. The minimum line width of the high-melting-point metal silicide film missing part detection element is the same as the minimum line width of the gate electrode 107 a. The side of the high-melting-point metal silicide film missing part detection element as well is covered directly by the insulation film spacer 122. In this second embodiment of the present invention as well, there is no restriction that the high-melting-point metal silicide film missing part detection element can be provided on the surface of the insulation film 104, and this can be alternately provided so as to directly cover the surface of the gate electrode film 105.
The polysilicon film pattern 111 is formed by an n+-type region 117 ab and a p+-type region 117 ba which are mutually connected. The lengths of the n+-type region 117 ab and the p+-type region 117 ba are, for example, approximately 0.5 μm, respectively. The film thickness of the cobalt silicide film pattern 131A as well is not uniform. The film thickness of the cobalt silicide film pattern 131A in parts above the n+-type region 117 ab is the same as the film thickness of the cobalt silicide film pattern 132 a, and the film thickness of the cobalt silicide film pattern 131A in parts above the p+-type region 117 ba is the same as the film thickness of the cobalt silicide film pattern 132 b. Directly above the junction part between the n+-type region 117 ab and the p+-type region 117 ba, it is possible for a missing part 133 to form in the cobalt silicide film pattern 131A.
A contact hole that reaches the two ends of the high-melting-point metal silicide film missing part detection element is provided in the interlayer insulation film 135. A pair of metal interconnects 138 connected to each of the ends of the high-melting-point metal silicide film missing part detection element is provided on the surface of the interlayer insulation film 135, within the contact holes. For this reason, the line widths at the two ends of the high-melting-point metal silicide film missing part detection element are wider than the minimum line widths thereof, this being, for example, approximately 0.5 μm. In the case in which it is necessary to detect the above-noted missing part 133 immediately before formation of the interlayer insulation film 135, the line width of both ends of the high-melting-point metal silicide film missing part detection element, because of the need to make connection to this part using a needle probe, is made, for example, 20 μm or larger.
The silicidation in the cobalt silicide, differing from the case of titanium silicide, is known to exhibit diffusion of cobalt within the silicon. The reason that the film thickness of the cobalt silicide film 132 b and cobalt silicide layer 132 ba is thinner than the film thickness of the cobalt silicide film 132 a and the cobalt silicide layer 132 aa is that because the ratio of occurrence of cobalt silicide seeds is greater in the n+ layer silicon surface than in the p+ layer silicon surface. The occurrence of a missing part 133 in the cobalt film silicide pattern 133A, similar to the case of a titanium silicide film pattern, is due to the existence of a difference between the film thickness of the cobalt silicide film pattern at the n+ layer surface and the film thickness of the cobalt silicide film pattern at the p+ layer surface. These missing parts 133, similar to the case of the first embodiment of the present invention, occur at the silicidation steps and thereafter, and further occur after the completion of the semiconductor device.
In the second embodiment of the present invention as well, similar to the first embodiment, in the case in which at least one missing part 133 occurs in the cobalt silicide film pattern 131A, if the resistance is measured by applying a bias between the pair of metal electrodes 138, a measured resistance value will be sufficiently larger than the simple sheet resistance value. Accordingly, by providing the above-noted high-melting-point metal silicide film missing part detection element formed by a polysilicon film pattern 111 and a cobalt silicide film pattern 131A in a semiconductor that includes a CMOS transistor having a salicide construction, it is possible to detect a minute missing part in a high-melting-point metal silicide film in a gate electrode made of a polycide interconnect.
A method for manufacturing the semiconductor device according to the second embodiment of the present invention is described below, making reference to FIG. 7, which is a cross-sectional viewshowing the major manufacturing process steps for a high-noting-point metal suicide film missing part detection element, and FIG. 6.
First, a p-type well 102, an n-type well 103, and a field insulation film 104 are formed on the surface of a silicon substrate 101, after that, a gate oxide film 105 having a thickness of 2.5 nm to 3.5 nm is formed by thermal oxidation processing.
Next, an undoped polysilicon film having a film thickness of 150 nm or so is formed over the entire surface, this undoped polysilicon film is patterned, and the undoped polysilicon film pattern for a high melting point silicide missing part detection element is formed on the surface of the field insulation film 104, for example. Simultaneously with this, by patterning an undoped polysilicon film, the gate electrodes for an n-channel MOS transistor and the p-channel MOS transistor are formed so as to extend from the surface of the p-type well 102 to the surface of the n-type well 103 passing through on the field insulation film 104. The minimum width of these undoped polysilicon film patterns is 0.18 μm. The line width at both ends of the undoped polysilicon film patterns for the high-melting-point metal silicide missing part detection element is made wider than this minimum width, for example, approximately 0.5 μm, and can be made 20 μm or greater, if necessary.
Next, a first photoresist film pattern having the same shape as the first photoresist film pattern 141 described above is formed on the substrate 101, this first photoresist film pattern is used as a mask to implant arsenic ions in the order of 1014 cm−2, thereby the undoped polysilicon film pattern in the part corresponding to the gate electrode of the n-channel MOS transistor becomes the n-type polysilicon film, simultaneously with which the n-type source/drain extension diffusion layer 121 a self-aligned to the n-type polysilicon film is formed on the p-type well 102 (FIG. 6(b)).
After removal of the above-noted first photoresist film pattern, a second photoresist film pattern having the same shape as the first photoresist film pattern 142 is formed. This second photoresist film pattern is used as a mask to perform ion implantation of boron (or boron fluoride) to the order of 1014 cm−2, the remaining part of the undoped polysilicon film pattern for the gate electrode becoming a p-type polysilicon film pattern, simultaneously with which a p-type source/drain extension diffusion layer 121 b is formed on the n-type well 103, self-aligned with this p-type polysilicon film pattern (FIG. 6(b)).
In this second embodiment as well, there is no particular restriction with regard to the sequence of the steps related to the first and second photoresist film patterns, and it is possible to reverse this sequence. If necessary, the first and second photoresist film patterns can be used as a mask to perform ion implantation of a p-type impurity and an n-type impurity, so that a p-type pocket diffusion layer and an n-type pocket diffusion layer are formed on the surfaces of the p-type well 102 and the n-type well 103, respectively.
Next, after removing the above-noted second photoresist film pattern, an insulation film (silicon oxide film or silicon nitride film) 130 is formed over the entire surface, using a LPCVD process, this insulation film being then etched back using an anisotropic etching, so as to form an insulation spacer 122 on the side surfaces of the undoped polysilicon film pattern for the above-noted high-melting-point metal silicide film missing part detection element and the n-type and p-type polysilicon patterns (FIG. 6(b).
Next, a third photoresist film pattern, having the same shape as the photoresist film pattern 143 in the first embodiment is formed. This third photoresist film pattern is used as a mask to perform ion implantation of arsenic to approximately 5×1015 cm−2, thereby the above-noted n-type polysilicon film pattern becoming an n+-type polysilicon film pattern, and an n+-type source/drain diffusion layer 123 aa being formed on the p-type well 102 surface, self-aligned with the n+-type polysilicon film pattern and insulation film spacer 122. Simultaneously with this, an n+-type region 117 ab is form on the undoped polysilicon film pattern for the above-noted high melting point metal silicide film missing part detection element. The depth of the junction of the n+-type source/drain diffusion layer 124 aa is, for example, 0.10 μm or less, and the length and the spacing of the n+-type region are, for example, less than 0.6 μm and greater than 0.4 μm, respectively (FIG. 6(b)).
Next, after removing the above-noted third photoresist film pattern, a fourth photoresist film pattern, having the same shape as the photoresist film pattern 144 of the first embodiment, is formed. This fourth photoresist film pattern is used as a mask to perform ion implantation of boron to approximately 5×1015 cm−2, using an energy of 2 keV, for example, so that the above-noted p-type polysilicon film pattern becomes a p+-type polysilicon film pattern and the p+-type source/drain diffusion layer is formed on the surface of the n-type well 103, self-aligned to the p+-type polysilicon film pattern and insulation film spacer 122. Simultaneously with this, in the part of the undoped polysilicon film pattern for the high-melting-point metal silicide film missing part detection element in which the above-noted n+-type region is not formed, a p+-type region 117 b is formed, the above-noted n+-type region becoming the n+-type region 117 aa. The depth of the junction of the p+-type source/drain diffusion layer is, for example, 0.15 μm or so, and the lengths of the n+-type region 117 aa and p+-type region 117 b are each, for example, approximately 0.5 μm (FIG. 6(a), FIG. 7(a)).
In this second embodiment as well, there is no particular restriction with regard to the sequence of the steps related to the third and fourth photoresist film patterns, and this sequence can be reversed.
Next, after removing the above-noted fourth photoresist film pattern, a cobalt film 130 is formed over the entire surface, using, for example, sputtering (FIG. 7(a)).
Next, a siliciding reaction is caused at a temperature of, for example, 900° C. By doing this, in the part that becomes the high-melting-point metal silicide film missing part detection element, the n+-type region 117 aa and p+-type region 117 b become an n+-type region 117 ab and p+-type region 117 ba, respectively, the original undoped polysilicon film pattern for the high-melting-point-metal silicide film missing part detection element becoming a polysilicon film pattern in which there is alternating connection between the n+-type region 117 ab and the p+-type region 117 ba, a cobalt silicide film pattern 131 being formed on the surface of the polysilicon film pattern 111, in a self-aligned manner.
In the part that becomes the n-channel MOS transistor, the above-noted n+ type polysilicon film pattern becomes an n+-type polysilicon film pattern 115 ab, and a cobalt silicide film pattern 132 a is formed on the surface of the n+-type polysilicon film pattern 115 ab, in self-aligned manner, thereby forming a gate electrode 107 a by these laminated semiconductor layers. Simultaneously with this, the above-noted n+-type source/drain diffusion layer becomes an n+-type source/drain diffusion layer 124 aa, and a cobalt silicide film 132 aa is formed on the n+-type source/drain diffusion layer 124 aa, in self-aligning manner, thereby forming a source/drain region 109 a formed by an n-type source/drain extension diffusion layer 121 a, an n+-type source/drain diffusion layer 124 aa, and a cobalt silicide film 132 aa.
In the part that becomes the p-channel MOS transistor, the above-noted p+-type polysilicon film pattern becomes a p+-type polysilicon film pattern 115 bb and a cobalt silicide film pattern 132 b is formed on the surface of the p+-type polysilicon film pattern 115 bb in self-aligned manner, these laminated semiconductor films forming a gate electrode 107 b. Simultaneously with this, the above-noted p+-type source/drain diffusion layer becomes a p+-type source/drain diffusion layer 124 ba, and a cobalt silicide film 132 ba is formed on the p+-type source/drain diffusion layer 124 ba, in self-aligned manner, thereby forming a source/drain region 109 b by the p-type source/drain extension diffusion layer 121 b, the p+-type source/drain diffusion layer 124 ba, and the cobalt silicide layer 132 ba (FIG. 6(b)).
Next, an unreacted titanium film 130 that directly covers the surfaces of the field insulation film 104 and the insulation film spacer 122 is selectively removed (FIG. 7(c)).
After the above, an interlayer insulation film 135 formed by a silicon oxide insulation film is formed over the entire surface, and a contact hole is formed in this interlayer insulation film 135. Additionally, metal interconnects 138, 139 a, 139 b, and 139 c are formed on the surface of the interlayer insulation film 135, thereby completing the semiconductor device of the second embodiment (FIG. 6).
Although the above-described first and second embodiments relate to a semiconductor device having a titanium salicide structure and a cobalt salicide structure, the first embodiment is not restricted in this manner, and can be applied as well to other high-melting-point metal salicide (for example, nickel). It will also be understood that the above-noted first and second embodiments, although described with regard to various specific values, are not restricted to those values.
In the first and second embodiments, the polysilicon film pattern of the high-melting point metal silicide film missing part detection element has alternating connection between an n+-type region and the p+-type region. However, the polysilicon film pattern of the high-melting-point metal silicide film missing part detection element of the present invention is not limited to this configuration.
Third Embodiment
The third embodiment of the present invention will be described with reference to FIG. 8(b) which is a schematic cross-sectional view of a CMOS transistor having a salicide structure, and FIG. 8(a) which is a schematic cross-sectional view of a high-melting-point metal silicide film missing part detection element. The third embodiment differs from the first embodiment in that the semiconductor device according to the third embodiment includes a high-melting-point metal silicide film missing part detection element, in which an n+-type region and p-type region are alternately connected in a polysilicon film pattern, the structure of this semiconductor being described further below.
A p-type well 202, n-type well 203, and a field insulation film 204 are provided on the surface of a silicon substrate 201. The element separation region provided on the surface of the silicon substrate 201 is formed, for example, as an STI structure, in which a shallow groove is filled by the field insulation film 204. The p-type well 202 and n-type well 203 are separated by this field insulation film 204. An n-channel MOS transistor is provided on the surface of the p-type well 202, and a p-channel MOS transistor is provided on the surface of the n-type well 203, the above-noted high-melting-point metal silicide film missing part detection element being formed, for example, on the surface of the field insulation film 204. The n-channel MOS transistor, the p-channel MOS transistor and the high-melting-point metal silicide film missing part detection element are covered by an interlayer insulation film 235.
The n-channel MOS transistor is formed by a gate oxide film 205, a gate electrode 206 a, provided on the surface of the p-type well 202, via the gate oxide film 205, and a source/drain region 208 a provided on the surface on the p-type well 202. The side surface of the gate electrode 206 a is directly covered by an insulation film spacer 222. The source/drain region 208 a is formed by an n-type source/drain extension diffusion layer 220 a provided on the surface of the p-type well 202, self-align to the polysilicon film pattern 213 ab, an n+-type source/drain diffusion layer 223 aaprovided on the surface of the p-type well 202, self-align to the gate electrode 206 a and insulation film spacer 222, and a titanium silicide layer 227 aa provided on the surface of the n+-type source/drain diffusion layer 223 aa, self-align with the gate electrode 206 a and insulation film spacer 222. The gate electrode 206 a is formed by an n+-type polysilicon film pattern 213 ab which directly covers the gate oxide film 205 and extends over the surface of the field insulation film 204, and a titanium silicide film pattern 227 provided on the surface of the n+-type polysilicon film pattern 213 ab in self-aligning manner.
The minimum line width of the gate electrode 206 a is, for example, 0.2 μm to 0.35 μm, the thicknesses of the titanium silicide film 227 a and the titanium silicide film 227 aa are approximately 270 nm to 450 nm, and the depth of the junction of the n-type source/drain diffusion layer 223 aa is, for example, approximately 0.15 μm. The n-type source/drain extension diffusion layer 220 a is formed by ion implantation of arsenic in the order of 1014 cm−2, and the n+-type source/drain diffusion layer 223 aa is formed by ion implantation of arsenic to approximately 2×1015 cm−2. The insulation film spacer 222 is formed by a silicon oxide film or a silicon nitride film having a thickness of, for example, 50 nm. The thickness of the gate oxide film is, for example, approximately 5 nm to 6 nm.
The p-channel MOS transistor is formed by the gate oxide film 205, the gate electrode 206 b provided on the surface of the n-type well 203 via the gate oxide film 205, and the source/drain region 208 b provided on the surface of then-type well 203. The source/drain region 208 b is formed by the p-type source/drain extension diffusion layer 220 b provided on the surface of the n-type well 203, self-align to the polysilicon film pattern 213 bb, the p+-type source/drain diffusion layer 223 ba provided on the surface of the n-type well 203, self-align to the gate electrode 206 b and the insulation film spacer 222, and the titanium silicide layer 227 ba provided on the surface of the p+-type source/drain diffusion layer 223 ba, self-align to the gate electrode 206 b and the insulation film spacer 222. The gate electrode 206 b is formed by the p+-type polysilicon film pattern 213 bb that directly covers the gate oxide film 205 and extends over the surface of the field insulation film 204, and the titanium silicide film pattern 227 b provided on the surface of the p+-type polysilicon pattern 213 bb in self-aligned manner.
The minimum line width of the gate electrode 206 b is the same as the minimum line width of the gate electrode 206 a. The film thickness of the titanium silicide film 227 b and titanium silicide film 227 ba is smaller than the film thickness of the titanium silicide film 227 a and titanium silicide film 227 aa by approximately 10%, this being approximately 300 nm to 500 nm. The depth of the junction of the p+-type source/drain diffusion layer 223 ba is, for example, approximately 0.2 μm. The p-type source/drain extension diffusion layer 220 b is formed by ion implantation of boron or boron fluoride to the order of 1014cm−2. The p+-type source/drain diffusion layer 223 ba is formed by ion implantation of boron to approximately 5×1015 cm−2, using an energy, for example, of 2 keV.
Contact holes are formed in the interlayer insulation film 235 so as to extend to the source/ drain regions 208 a and 208 b. On the surface of the interlayer insulation film 235, via these contact holes, are formed, for example, a metal interconnect 237 a connected to one side of the source/drain region 208 a, a metal interconnect 237 b connected to one side of the source/drain region 208 b, and a metal interconnect 237 cconnected to the other source/drain region 208 a and the other source/drain region 207 b. For example, the metal interconnect 237 a is connected to the power supply voltage, and the metal interconnect 237 b is connected to the ground voltage.
A high-melting-point metal silicide film missing part detection element is formed, for example, by a polysilicon film pattern so as to directly cover the surface of a field insulation film 204, and a titanium silicide film pattern 226A provided on the surface of the polysilicon film pattern 210 in self-aligning manner. The minimum line width of the high-melting-point metal silicide film missing part detection element is the same as the minimum line width of the gate electrode 206 a. The side surface of the high-melting-point metal silicide film missing point detection element as well is directly covered by the insulation film spacer 222. In this embodiment as well, the high-melting-point metal silicide film missing part detection element is not restricted to being provided on the surface of the field insulation film 204, and can alternately be provided, for example, so as to directly cover the surface of the gate oxide film 205.
The polysilicon film pattern 210, in contrast to the above-noted first embodiment, is formed by an alternate connection between the n+-type region 216 aa and p-type region 218 ba. The film thickness of the titanium silicide film pattern 226A is not uniform, but rather has a thickness that is the same of the film thickness of the titanium silicide film pattern 227 a in the part above the surface of the n+-type region 216 aa, and is thicker than the titanium silicide film 227 b in the part on the surface of the p-type region 218 ba. Directly above the junction part between the n+-type region 216 aa and p-type region 218 ba a missing part 228 can occur in the titanium silicide film pattern 226A.
Contact holes that reach the two end parts of the high-melting-point metal silicide film missing part detection element are provided in the interlayer insulation film 235. A pair of metal interconnects 236 which are connected respectively to the end part of the high-melting-point metal silicide film missing part detection element are provided on the surface of the interlayer insulation film 235, via these contact holes. For this reason, the line width at the two ends of the high-melting-point metal silicide film missing part detection element are both wider than this minimum line length (for example, approximately 0.5 μm). Additionally, the two end parts of the polysilicon film pattern 216 are formed by the n+-type region 216 aa. In the case in which it is necessary to detect the above-noted missing part 288 immediately before formation of the interlayer insulation film 235, the line width of both ends of the high-melting-point metal silicide film missing part detection element, because of the need to make connection to this part using a needle probe, is made, for example, 20 μm or larger.
In this embodiment of the present invention as well, similar to the case of the first embodiment, in a semiconductor device including a CMOS transistor having a salicide structure, the above-noted high-melting-point metal silicide film missing part detection element, formed by a polysilicon film pattern 210 and a titanium silicide film pattern 226A is provided, so that it easy to detect a minute missing part in the high-melting-point metal silicide film in the gate electrode made of a polycide interconnect.
Additionally, in this embodiment of the present invention, as described above, the ratio of (thickness of the titanium silicide film pattern 226A directly above the p-type region 218 ba)/(thickness of the titanium silicide film pattern 226A directly above the n+-type region 216 aa) is a greater value than the ratio of (thickness of the titanium silicide pattern 126A directly above the p+-type region 116 ba)/(thickness of the titanium silicide film pattern 126A directly above the n+-type region 116 ab). That is, the rate of occurrence of the missing part 228 is greater than that of the missing part 128. As a result, the detection rate of missing parts in the high-melting-point metal silicide film of the gate electrode in this example of the present invention is higher than the first embodiment of the present invention.
Although the above-noted embodiment of the present invention relates to a semiconductor device having a titanium salicide structure, the third embodiment is not restricted in this manner, and the third embodiment can be similarly applied to the case of a different high-melting-point metal salicide in which the film thickness of the high-melting-point metal silicide on the n+-type silicon surface when silicidation occurs is smaller (thinner) than the film thickness of the high-melting-point metal silicide film on formed the p-type silicon surface. It will also be understood that the various values presented in the above-noted description of the example of the third embodiment of the present invention are not restricted to those values.
This third embodiment of the present invention will be described further below, with reference to the FIG. 9 and FIG. 10, which are schematic cross-sectional views of the manufacturing process or high-melting-point metal silicide missing part detection element, FIG. 11 and FIG. 12, which are schematic cross-sectional views showing a manufacturing process for a titanium salicide structure CMOS transistor, and the above-noted FIG. 8.
First, a p-type well 202, and n-type well 203, and a field insulation film 204 made from a silicon oxide insulation film are formed on the surface of a silicon substrate 201. A gate oxide film 205 having a thickness of, for example, 5 nm to 6 nm is formed on the surface of the p-type well 202 and n-type well 203, using thermal oxidation.
Next, a process such as low-pressure CVD is used to form an undoped polysilicon film (not shown in the drawing) to a thickness of approximately 150 nm to 200 nm over the entire surface of the substrate 201, this undoped polysilicon film being then patterned and an undoped polysilicon film pattern 212 of the high-melting-point metal silicide film missing part detection element and undoped polysilicon film pattern 213 for a gate electrode being formed. The minimum line width of the undoped polysilicon film pattern 212 is the same as the minimum line width of the undoped polysilicon film pattern 213, this being for example 0.2 μm to 0.35 μm. The line width of the two ends of the undoped polysilicon film 212 are wider than this minimum line width, this being, for example, approximately 0.5 μm, which can be made to 5 μm or greater, if necessary.
Next, a first photoresist film pattern 241, having an aperture at a part corresponding to the p-type well 202, and covering the undoped polysilicon film pattern 212 and at least the surface of the n-type well 203 is formed. This photoresist film pattern 241 is used as a mask in performing ion implantation of arsenic to the order of 1014 cm−2, thereby a part of the undoped polysilicon film pattern 213 becoming the n-type polysilicon film pattern 213 a simultaneously with which an n-type source/drain extension diffusion layer 220 a is formed on the surface of the p-type well 202, self-aligned to the n-type polysilicon film pattern 213 a (FIG. 9(a), FIG. 11(a)).
After removing the above-noted photoresist film pattern 241, a second photoresist film pattern 242 is formed, the photoresist film pattern 242 has apertures at a part corresponding to the n-type well 203 and the part corresponding to the undoped polysilicon film pattern 212, and minimally covers the surface of the p-type well 202. This photoresist film pattern 242 is used as a mask in performing ion implantation of boron (or boron fluoride) to the order of 1014 cm−2, the remaining part of the undoped polysilicon film pattern 213 becoming the p-type polysilicon film pattern 213 b, simultaneously with which a p-type source/drain extension diffusion region 220 b is formed on the surface of the n-type well 203, self-align to the p-type polysilicon film pattern 213 b, and simultaneously with which the undoped polysilicon film pattern 212 becomes the p-type polysilicon film pattern 212 b (FIG. 9(b), FIG. 11(b)).
In this embodiment of the present invention as well, there is no restriction to the sequence of the process steps related to the photoresist film patterns 241 and 242, and it is possible to reverse this sequence. Additionally, if necessary it is possible to use the photoresist film patterns 241 and 242 as mask for ion implantation of a p-type impurity and an n-type impurity, so as to form a p-type pocket diffusion layer and n-type pocket diffusion layer, respectively, on the surfaces of the p-type well 202 and n-type well 203.
Next, the photoresist film pattern 242 is removed, and insulation film (made of silicon oxide or silicon nitride) is formed on the entire surface to a film thickness of, for example, 50 nm, using a low-pressure CVD process. This insulation film is etched back using anisotropic edging, so as to form insulation film spacer 222 on the side surfaces of the p-type polysilicon film pattern 212 b, n-type polysilicon film pattern 213 a, and p-type polysilicon film pattern 213 b.
Next, a third photoresist film pattern 243 is formed. The photoresist pattern 243 has an aperture in a part corresponding to the p-type well 202 and minimally covers the surface of the n-type well 203. Additionally, the photoresist film pattern 243 is provided with a plurality of apertures in a part that covers the p-type polysilicon film pattern 212 b. The aperture width and aperture spacing in the part that covers the p-type polysilicon film pattern 212 b (a part other than the end part of the p-type polysilicon film pattern 212 b), are, respectively, approximately 0.4 μm and 0.6 μm. This photoresist film pattern 243 is used as a mask, for example, to perform ion implantation of arsenic to approximately 2×1015 cm−2, thereby the n-type polysilicon film pattern 213 a becoming the n+-type polysilicon film pattern 213 aa and an n+-type source/drain diffusion layer 223 abeing formed on the surface of the p-type well 202, self-aligned with the n+-type polysilicon film pattern 213 aa and insulation film spacer 222. Simultaneously with this, an n+-type region 216 a is formed on the p-type polysilicon film pattern 212 b, the remaining part becoming the p-type region 218 b. The depth of the junction of the n+-type source/drain diffusion layer 223 a is, for example, approximately 0.15 μm, and the length and spacing of the n+-type region 216 a are approximately 0.5 μm (FIG. 9(c), FIG. 11(c)).
Next, after removing the above-noted photoresist film pattern 243 a fourth photoresist film pattern 244 is formed. The photoresist film pattern 244 has an aperture in a part corresponding to the n-type well 203, and covers a polysilicon film pattern having alternate connection between the n+-type region 216 a and p-type region 218 b and minimally the surface of the p-type well 202. The photoresist film pattern 244 is used as a mask, for example, in performing ion implantation of boron to approximately 5×1015 cm−2 using an energy of 2 keV, thereby the p-type polysilicon film pattern 213 b becoming the p+-type polysilicon film pattern 213 ba and a p+-type source/drain diffusion layer 223 b being formed on the surface of the n-type well 203, self-aligned to the p+-type polysilicon film pattern 213 ba and the insulation film spacer 222. The depth of the junction of the p+-type source/drain diffusion layer 223 b is, for example, approximately 0.2 μm (FIG. 9(d), FIG. 11(d)). In these example of the present invention as well, there is no restriction with regards to the sequence of the process steps related to the photoresist film pattern 243 and 244, which can be reversed.
Next, after removing the above-noted photoresist film pattern 244, sputtering or the like is used to form a titanium film 225 over the entire surface (FIG. 10(a), FIG. 12(a)).
A siliciding reaction is cause to occur at a prescribed temperature. By doing this, in a part that becomes the high-melting-point metal silicide film missing part detection element, the n+-type region 216 a and p-type region 218 b become, respectively, the n+-type region 216 aa and p-type region 218 ba, accordingly the original undoped polysilicon film pattern 212 becoming a polysilicon film pattern 210, having alternate connection between the n+-type region 216 aa and p-type region 218 ba, with a titanium silicide film pattern 226 formed on the surface of the polysilicon film pattern 210 in self-aligning manner.
In the part that becomes the n-channel MOS transistor, the n+-type polysilicon film pattern 213 aa becomes the n+-type polysilicon film pattern 213 ab, and a titanium silicide film pattern 227 a is formed in self-aligned manner on the surface of the n+-type polysilicon film pattern 213 ab, these laminated semiconductor layers forming the gate electrode 206 a.
Simultaneously with this, the n+-type source/drain diffusion layer 223 a becomes the n+-type source/drain diffusion layer 223 aa, and the titanium silicide layer 227 aa is formed in self-aligned manner on the n+-type source/drain diffusion layer 223 aa, thereby forming a source/drain region 208 a by the n-type source/drain extension diffusion layer 220 a, the n+-type source/drain diffusion layer 223 aa and the titanium silicide layer 227 aa.
In the part that becomes the p-channel MOS transistor, the p+-type polysilicon film pattern 213 ba becomes the p+-type polysilicon film pattern 213 bb and a titanium silicide film pattern 227 b is formed in self-aligning manner on the surface of the p+-type polysilicon film pattern 213 bb, these laminated semiconductor layers forming the gate electrode 206 b. Simultaneously with this the p+-type source/drain diffusion layer 223 b becomes the p+-type source/drain diffusion layer 223 ba and a titanium silicide layer 227 ba is formed in self-aligning manner on the surface of the p+-type source/drain diffusion layer 223 ba, thereby forming a source/drain region 208 b that is formed by the p-type source/drain extension diffusion layer 220 b, the p+-type source/drain diffusion layer 223 ba and the titanium silicide layer 227 ba (FIG. 10(b), FIG. 12(b)).
In the above described titanium siliciding reaction, it is known that silicon diffuses into the titanium. If boron is present in high concentration in the silicon, the siliciding reaction speed will become low. Additionally, in this reaction the occurrence of titanium silicide kernels is higher in a p-type layer silicon surface having a low impurity concentration than in a p+-type layer silicon surface. As a result, the film thickness of the titanium silicide film pattern 226 directly above the n+-type region 216 aa is thinner than that of the titanium silicide film pattern 226 directly above the p-type region 218 ba.
Next, the unreacted titanium film which directly covers the surface of the field insulation film 204 and insulation film spacer 222 is selectively removed (FIG. 10(c), FIG. 12(c)).
After the above, an interlayer insulation film 235, formed by a silicon oxide insulation film, is formed over the entire surface, and contact holes are formed in this interlayer insulation film. Additionally, metal interconnects 236, 237 a, 237 b, and 237 c are formed in the surface of the interlayer insulation film 235, thereby completing the semiconductor device according to the third embodiment of the present invention (FIG. 8).
During the manufacturing process including the above-noted siliciding reaction, and after the completion of the semiconductor device, in the titanium silicide film pattern 226 that forms the high-melting-point metal silicide film missing part detection element, a missing part 228 occurs at a position corresponding to a position directly above the jointing part between the n+-type region 216 aa and p-type region 218 b in the polysilicon film pattern 210, so that the titanium silicide film pattern 226 becomes the titanium silicide pattern 226A. The occurrence of the missing part 228 is due to the difference in film thicknesses between the film thickness of the titanium silicide film pattern above the n+-type region 216 aa and the film thickness of the titanium silicide pattern film above the p-type region 218 ba.
Additionally coagulation of the titanium silicide film occurred in the phase transition is also a cause of the tendency for the above-noted missing part 228 to occur.
Fourth Embodiment
The fourth embodiment of the present invention will be described with reference to FIG. 13(b) which is a schematic cross-sectional view of a CMOS transistor having a salicide structure, and FIG. 13(a) which is a schematic cross-sectional view of a high-melting-point metal silicide film missing part detection element. The fourth embodiment differs from the second embodiment in that the semiconductor device according to the fourth embodiment includes a high-melting-point metal silicide film missing part detection element, in which an p+-type region and n-type region are alternately connected in a polysilicon film pattern, the structure of this semiconductor being described further below.
In this semiconductor device, a p-type well 302, an n-type well 303, and a field insulation film 304 are provided on the surface of a silicon substrate 301. An n-channel MOS transistor provided on the surface of the p-type well 302 and a p-channel MOS transistor provided on the surface of the n-type well 303 are provided on the surface of the n-channel well 303, the above-noted high-melting-point metal silicide film missing part detection element being provided, for example, on the surface of the field insulation film 304. The n-channel MOS transistor, p-channel MOS transistor, and high-melting-point metal silicide film missing part detection element are covered by an interlayer insulation film 335.
The n-channel MOS transistor is formed by a gate oxide film 305, a gate electrode 307 a provided on the surface of the p-type well 302 via the gate oxide film 305, and a source/drain region 309 a provided on the surface of the p-type well 302. The side surface of the gate electrode 307 a is directly covered by an insulation film spacer 322. The source/drain region 309 a is formed by an n-type source/drain extension diffusion region 321 a provided on the surface of the p-type well 302, self-align to the polysilicon film pattern 315 a, an n+-type source/drain diffusion layer 324 aa provide on the surface of the p-type well 302, self-aligned to the gate electrode 307 a and the insulation film spacer 322, and a cobalt silicide layer 332 aa provided on the surface of the n+-type source/drain diffusion layer 324 aa, self-aligned to the gate electrode 307 a and insulation film spacer 322. The gate electrode 307 a is formed by a n+-type polysilicon film pattern 315 ab that directly covers the surface of the gate oxide film 305 and extends over the surface of the field insulation film 304, and a cobalt silicide film pattern 332 a formed on the surface of the n-type polysilicon film pattern 315 ab, in self-aligning manner.
The minimum line width of the gate electrode 307 a is, for example, 0.18 μm and less, the thicknesses of the cobalt silicide film 332 a and the cobalt silicide film 332 aa are approximately 250 nm to 350 nm, and the depth of the junction of the n+-type source/drain diffusion layer 324 aa is, for example, approximately 0.1 μm. The n-type source/drain extension diffusion layer 321 a is formed by ion implantation of arsenic in the order of 1014 cm−2, and the n+-type source/drain diffusion layer 324 aa is formed by ion implantation of arsenic to approximately 5×1015 cm−2. The insulation film spacer 322 is formed by a silicon oxide film or a silicon nitride film having a thickness of, for example, 50 nm. The thickness of the gate oxide film 305 is, for example, approximately 2.5 nm to 3.5 nm.
The p-channel MOS transistor is formed by the gate oxide film 305, the gate electrode 307 b provided on the surface of the n-type well 303 via the gate oxide film 305, and the source/drain region 309 b provided on the surface of the n-type well 303. The source/drain region 309 b is formed by the p-type source/drain extension diffusion layer 321 b provided on the surface of the n-type well 303, self-aligned to the polysilicon film pattern 315 b, the p+-type source/drain diffusion layer 324 ba provided on the surface of the n-type well 303, self-aligned to the gate electrode 307 b and the insulation film spacer 322, and the cobalt silicide layer 332 ba provided on the surface of the p+-type source/drain diffusion layer 324 ba, self-aligned to the gate electrode 307 b and the insulation film spacer 322. The gate electrode 307 b is formed by the p+-type polysilicon film pattern 315 bb that directly covers the gate oxide film 305 and extends over the surface of the field insulation film 304, and the cobalt silicide film pattern 332 b provided on the surface of the p+-type polysilicon pattern 315 bb in self-aligned manner.
The minimum line width of the gate electrode 307 b is the same as the minimum line width of the gate electrode 307 a. The film thickness of the cobalt silicide film 332 b and the cobalt silicide layer 332 ba is approximately 5% thinner than the film thicknesses of the cobalt silicide film 332 a and cobalt silicide layer 332 aa, respectively, these film thicknesses are, for example, approximately 240 nm to 330 nm.
The depth of the junction of the p+-type source/drain diffusion layer 324 ba is, for example, approximately 0.15 μm. The p-type source/drain extension diffusion layer 321 b is formed by ion implantation of either boron of boron fluoride to the order of 1014 cm−2. The p+-type source/drain diffusion layer 324 ba is formed by ion implantation of boron to approximately 2×1015 cm−2 to 3×1015 cm−2, with an energy, for example, 2 keV.
Contact holes are provided in the interlayer insulation film 335, so as to reach the source/ drain regions 309 a and 309 b. In the surface of the interlayer insulation film 335, a metal interconnect 339 a connected to one side of the source/drain region 309 a, a metal interconnect connected to one side of the source/drain region 309 b, and a metal interconnect 339 c connected to the other source/drain region 309 a and the other source/drain region 309 b are formed, via these contact holes. For example, the metal interconnect 339 a is connected to a power supply voltage, and a metal interconnect 339 b is connected to a ground voltage.
The high-melting-point metal silicide film missing part detection element is formed, for example, by a polysilicon film pattern 311 that directly covers the surface of the field insulation film 304, and a cobalt silicide film pattern 331A provided on the surface of the polysilicon film pattern 311 formed in self-aligned manner. The minimum line width of the high-melting-point metal silicide film missing part detection element is the same as the minimum line width of the gate electrode 307 a. The side surfaces of the high-melting-point metal silicide film missing part detection element as well are directly covered by the insulation film spacer 322. In this example of the present invention as well, there is no restriction to the high-melting-point metal silicide film missing part detection element being formed on the surface of the field insulation film 304, and it is possible, for example, to have a configuration in which there is direct covering of the gate oxide film 305.
The polysilicon film pattern 311 is formed by the alternating connection between the n-type region 319 aa and the p-type region 317 ba. The length of the n-type 319 aa and the p-type region 317 ba, are approximately 0.5 μm, respectively. The film thickness of the cobalt silicide film pattern 331A in this case as well is not uniform. The thickness of the part above the n-type region 319 aa is thinner than the film thickness of the cobalt silicide film pattern 332 a, and the thickness of the part above the P+-type region 317 ba is the same thickness of the cobalt silicide film pattern 332 b. Directly above the junction part between the n-type region 319 aa and the p+-type region 317 ba, a missing part 333 occurs in the cobalt silicide film pattern 331A.
The interlayer insulation film 335 is provided with a contact hole that extends to the two end parts of the high-melting point metal silicide film missing part detection element. A pair of metal interconnects 338 are provided on the surface of the interlayer insulation film 335, so as to make connection to the end parts of the high-melting point metal silicide film missing part detection element. For this reason, the line width of the two ends of the high-melting point metal silicide film missing part detection element are wider than the minimum line width thereof (for example, approximately 0.5 μm). Additionally, the two end pars of the polysilicon film pattern 310 are each p+-type regions 317 ba. In the case in which the above-noted missing part 333 is to be detected immediately before formation of the interlayer insulation film 335, the line width at the two end parts of the high-melting point metal silicide film missing part detection element are made 20 μm, for example, because of the need to make contact with these parts using a needle probe.
The silidation in the cobalt silicide, in contrast to the case of titanium silicide, is known to exhibit cobalt diffusing into the silicon. The reason that the thicknesses of the cobalt silicide film pattern 331A at a portion directly above the n-type region 319 aa, the cobalt silicide film 332 a, the cobalt silicide film 332 b decrease in this sequence is because the ratio of occurrence of cobalt silicide seeds decreases in the sequence of the n-type layer, n+-type layer, and p+-type layer. The reason that the missing part 333 occurs in the cobalt silicide film pattern 331A is due to the difference between the film thickness of the cobalt silicide film pattern 331A at the n-type region 319 aa and the film thickness of the cobalt silicide film pattern 331A at the p+-type region 317 ba, similar to the case of a titanium silicide pattern. These missing parts 333, similar to the case of the first embodiment of the present invention, occur at the silicidation steps and thereafter, and further occur after the completion of the semiconductor device.
This embodiment of the present invention achieves the effects of the second embodiment.
Additionally, in this embodiment, the ratio of (film thickness of cobalt silicide film pattern 331A directly above the n-type region 319 aa)/(film thickness of cobalt silicide film pattern 331A directly above the p+-type region 317 ba) becomes greater than the ratio of (film thickness of cobalt silicide film pattern 131A directly above the n+-type region 117 ab)/(film thickness of cobalt silicide film pattern 131A directly above the p+-type region 117 ba) (shown in the second embodiment). As a result, the detection rate of missing parts in the high-melting-point metal silicide film of the gate electrode in this embodiment is higher than the case of the second embodiment.
The manufacturing process for the fourth embodiment of the present invention is described below, with references being made to FIG. 14 and FIG. 15, which are schematic cross-sectional views showing the manufacturing process for the high-melting point metal silicide film missing part detection element, FIG. 16 and FIG. 17, which are schematic cross-sectional views showing the manufacturing process for the CMOS transistor having a cobalt salicide structure, and the above-noted FIG. 13.
First, a p-type well 302, an n-type well 303, and a field insulation film 304 made of a silicon oxide insulation film are formed on the surface of a silicon substrate 301. A gate oxide film 305, having a film thickness of, for example, 2.5 to 3.5 nm, is formed on the surface of the p-type well 302 and the n-type well 303.
Next, an undoped polysilicon film to a thickness of approximately 150 nm is formed on the entire surface, and this undoped polysilicon film is patterned, thereby forming an undoped polysilicon film pattern for a high-melting point metal silicide film missing part detection element on, for example, the field insulation film 304. Simultaneously, the undoped polysilicon film pattern 315 for the gate electrode on the gate electrode oxide 305 is formed, for example, extending from the surface of the p-type well 302, via the surface of the field insulation film 304, to the surface of the n-type well 303. The minimum line widths of these undoped polysilicon film patterns is, for example, less than 0.18 μm. The line width at both end parts of the undoped polysilicon film pattern for the high-melting point metal silicide film missing part detection element is made wider than this minimum line width, this being for example approximately 0.5 μm, and 20 μm or greater, if necessary.
Next, a first photoresist film pattern 341 is formed on the substrate 301. The photoresist film pattern 341 has an aperture in the surface of the p-type well 302 and an undoped polysilicon film pattern part for the high-melting point metal silicide film missing part detection element, the first photoresist film pattern 341 minimally covering the surface of the n-type well 303. The photoresist film pattern 341 is used as a mask to perform ion implantation of arsenic to the order of 1014 cm−2, thereby the undoped polysilicon film pattern 315 of a part corresponding to the gate electrode of the n-channel MOS transistor becoming a n-type polysilicon film pattern 315 a, simultaneously with which an n-type source/drain extension diffusion layer 321 a is formed on the surface of the p-type well 302, self-aligned with the n-type polysilicon film pattern 315 a, and the above-noted undoped polysilicon film pattern for the high-melting point metal silicide film missing part detection element becoming the n-type polysilicon film pattern 314 a (FIG. 14(a) and FIG. 16(a)).
After removal of the above-noted photoresist film pattern 341, a second photoresist film pattern 342 which has the same shape as the photoresist film pattern 142 used in the first embodiment, is formed. This photoresist film pattern 142 is used as a mask to perform ion implantation of boron (or boron fluoride) to the order of 1014 cm−2, so that the remaining part of the undoped polysilicon film pattern 315 becomes a p-type polysilicon film pattern 315 b, simultaneously with which a p-type source/drain extension diffusion layer 321 b is formed on the surface of the n-type well 303, self-aligned with the p-type polysilicon film pattern 315 b (FIG. 14(b) and FIG. 16(b)).
In this embodiment as well, there is no restriction with regard to the sequence of the steps relating to the photoresist film patterns 341 and 342, which can be reversed. If necessary, it is possible to perform ion implantation of a p-type impurity and n-type impurity using the photoresist film patterns 341 and 342 as masks, so as to form on the surfaces of the p-type well 302 and the n-type well 303 a p-type pocket diffusion layer and an n-type pocket diffusion layer, respectively.
Next, after removal of the above-noted photoresist film pattern 342, a silicon oxide or silicon nitride film is formed over the entire surface, using an LPCVD process, this insulation film being etched back using an anisotropic etching process, so that an insulation film spacer 322 is formed on the side surfaces of the n-type polysilicon film patterns 314 a and 315 a and the p-type polysilicon film pattern 315 b.
Next, a third photoresist film pattern 343 is formed. This photoresist film pattern 343 has an aperture at a part above the surface of the p-type well 302, covers the above-noted n-type polysilicon film pattern 314 a, and minimally covers the n-type well 303 surface. This photoresist film pattern 343 is used as a mask to perform ion implantation of arsenic to approximately 5×1015 cm−2, so that the n-type polysilicon film pattern 315 a becomes an n+-type polysilicon film pattern 315 aa and an n+-type source/drain diffusion layer 324 a is formed, self-aligned to the n+-type polysilicon film pattern 315 aa and the insulation spacer 322. The depth of the junction of the n+-type source/drain diffusion layer 324 a is, for example, approximately 0.10 μm and less (FIG. 14(c) and FIG. 16(c)).
Next, after removal of the photoresist film pattern 343, a fourth photoresist film pattern 344 is formed. This photoresist film pattern 344 has an aperture over the surface of the n-type well 303, minimally covers the p-type well 302 surface, and further has a plurality of apertures in parts that cove the n-type polysilicon film pattern 314 a. The aperture width and spacing of the apertures in the part that covers the n-type polysilicon film pattern 314 a, except for the end part of the n-type polysilicon film pattern 314 a) are approximately 0.4 μm and 0.6 μm, respectively. The photoresist film pattern 344 is used as a mask to perform ion implantation of boron to approximately 5×1015 cm−2 using, for example, an energy of 2 keV, so that the above-noted p-type polysilicon film pattern 315 b becomes a p+-type polysilicon film pattern 315 ba, a p+-type source/drain diffusion region 324 b being formed on the surface of the n-type well 303, self-aligned to the p+-type polysilicon film pattern 315 ba and the insulation film spacer 322. Simultaneously with this, a p+-type region 317 b is formed on the n-type polysilicon film pattern 314 a. The depth of the junction of the p+-type source/drain diffusion region 324 b is, for example, approximately 0.15 μm, and the lengths of the n-type region 319 a and p+-type region 317 b are each approximately 0.5 μm (FIG. 14(d) and FIG. 16(d)).
In this embodiment as well, there is no restriction with respect to the sequence of the process steps related to the photoresist film patterns 343 and 344, which can be reversed.
Next, after removing the above-noted photoresist film pattern 344, a cobalt film 330 is formed over the entire surface by, for example, a sputtering process (FIG. 15(a) and FIG. 17(a)).
Next, a siliciding reaction is caused at a temperature of, for example, 900° C. By doing this, in the part that becomes the high-melting-point metal silicide film missing part detection element, the n-type region 319 a and p+-type region 317 b become an n-type region 319 aa and p+-type region 317 ba, the original undoped polysilicon film pattern for the high-melting-point-metal silicide film missing part detection element becoming a polysilicon film pattern 311 in which there is alternating connection between the n-type region 319 aa and the p+-type region 317 ba, a cobalt silicide film pattern 331 being formed on the surface of the polysilicon film pattern 311, in a self-aligned manner.
In the part that becomes the n-channel MOS transistor, the above-noted n+ type polysilicon film pattern 315 aa becomes an n+-type polysilicon film pattern 315 ab, and a cobalt silicide film pattern 332 a is formed on the surface of the n+-type polysilicon film pattern 315 ab, in a self-aligned manner, thereby forming a gate electrode 307 a by these laminated semiconductor layers. Simultaneously with this, the above-noted n+-type source/drain diffusion layer 324 a becomes an n+-type source/drain diffusion layer 324 aa, and a cobalt silicide film 332 aa formed on the n+-type source/drain diffusion layer 324 aa, in a self-aligning manner, thereby forming a source/drain region 309 a formed by an n-type source/drain extension diffusion layer 321 a, an n+-type source/drain diffusion layer 324 aa, and a cobalt silicide film 332 aa.
In the part that becomes the p-channel MOS transistor, the above-noted p+-type polysilicon film pattern 315 ba becomes an n+-type polysilicon film pattern 315 bb and a cobalt silicide film pattern 332 b is formed on the surface of the p+-type polysilicon film pattern 315 bb in a self-aligned manner, these laminated semiconductor films forming a gate electrode 307 b. Simultaneously with this, the above-noted p+-type source/drain diffusion layer 324 b becomes a p+-type source/drain diffusion layer 324 ba, and a cobalt silicide film 332 ba formed on the p+-type source/drain diffusion layer 324 ba, in a self-aligned manner, thereby forming a source/drain region 309 b by the p-type source/drain extension diffusion layer 321 b, the p+-type source/drain diffusion layer 324 ba, and the cobalt silicide layer 332 ba (FIG. 15(a), FIG. 17(b)).
Next, an unreacted titanium film 330 that directly covers the surfaces of the field insulation film 304 and the insulation film spacer 322 is selectively removed (FIG. 15(c) and FIG. 17(c)).
After the above, an interlayer insulation film 335 formed by a silicon oxide insulation film is formed over the entire surface, and a contact hole is formed in this interlayer insulation film. Additionally, metal interconnects 338, 339 a, 339 b, and 339 c are formed on the surface of the interlayer insulation film 335, thereby completing the semiconductor device of the fourth embodiment (FIG. 13(a) and FIG. 13(b)).
Although the above-described fourth embodiment relates to a semiconductor device having a cobalt salicide structure, the fourth embodiment is not restricted in this manner, and can be applied as well to other high-melting-point metal salicide, in which the film thickness of the high-melting point metal silicide film directly above the p+-type region being thinner than the film thickness of the high-melting point metal silicide film directly above the n-type region. It will also be understood that the above-noted embodiment of the fourth embodiment, although described with regard to various specific values, is not restricted to those values.
As described in detail above, a polysilicon film pattern made up of a high-melting point metal silicide film missing part detection element according to the present invention is formed by alternate connections of either n+-type and p+-type regions, n+-type and p-type regions, or n-type and p+-type regions. By doing this, a film thickness in a part that is directly above the n+-type region or n-type region is different from the film thickness in apart that is directly above the p+-type region or p-type region.
As a result, there is a tendency for a missing part to occur in the high-melting point metal silicide film missing part detection element directly above the junction part between the n+-type region or n-type region and the p+-type region or p-type region, thereby facilitating fine detection formed in a gate electrode having a polycide structure.